Design And Development Of Substation Inspection Robot Control System

With the rapid development of power system scale and the maturity of robot technology,the use of patrol robots instead of manual patrol can effectively improve the efficiency of patrol,and gradually become the development trend of intelligent management of power grid.In this paper,a robot control system for substation patrol inspection is designed and developed according to the requirements of substation patrol inspection environment and task.The main research work is as follows:Firstly,the overall structure design and hardware integration scheme of inspection robot system are introduced.On this basis,an efficient path planning method for substation inspection is designed.The topological map of substation is established,and the path planning problem is decomposed into local path planning and global path planning.Dijkstra algorithm is used to find the local shortest path between any two inspection points in the inspection task.Then,aiming at shorter path and fewer passing points,simulated annealing algorithm and ant colony algorithm are combined to improve,and the relative optimal global path is obtained.The annealing algorithm receives the suboptimal solution with a certain probability,which improves the premature condition of ant colony algorithm.By using deviation degree to guide pheromone updating,the convergence speed is accelerated and the efficiency of global planning is improved.Secondly,based on the requirement of the patrol robot’s motion capability and the motion structure of four-wheel omni-directional full-drive,the kinematics model of the front-wheel steering and the rear-wheel driving of the robot and the in-situ steering model are established when the wheel axle center does not coincide with the reference position of the robot.The trajectory tracking control law is designed,and the global asymptotic stability of the system is analyzed.The desired trajectory is accurately tracked by the robot.Thirdly,aiming at the problem that the attitude estimation error of odometer is large when the driving wheel of robot is skidding,an attitude correction algorithm based on monocular vision is proposed.Based on the basic model of the camera,feature extraction and matching optimization are carried out with ORB features,and the relative pose calculation method is deduced according to the position relationship between the robot and the camera.The weighted fusion algorithm is used to fuse odometer attitude and visual attitude,which reduces the error of attitude estimation.Fourthly,according to the functional requirements of the system,the software of robot control system is developed.The software framework of the system is designed,and the functional modules are implemented based on the ROS platform.The test results of the actual robot system in the field environment prove that the substation inspection robot system designed in this paper can complete the inspection task steadily and reliably.